An AI-based iron smelting material proportioning monitoring and adjusting system

Abstract

The application relates to the technical field of iron smelting material proportioning, in particular to an AI-based iron smelting material proportioning monitoring and adjusting system. The system comprises a central control unit, which is used for setting a plurality of monitoring points; a material unit, which is used for collecting real-time charging material parameters; and a monitoring unit, which comprises a plurality of monitoring sub-modules and is arranged at the monitoring points and used for collecting the operating parameters of a direct current electric arc furnace. An AI agent and a plurality of iron smelting periods are established based on the central control unit; the AI agent generates an optimal material proportioning strategy according to the iron smelting material parameters in a single iron smelting period, sets corresponding charging parameters according to the material proportioning strategy, avoids uneven mixing of materials, simultaneously prewarns and corrects the charging deviation in a timely manner by monitoring the real-time charging amount, and avoids the influence of material proportioning on the iron smelting quality.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of iron smelting material proportioning, in particular to an AI-based iron smelting material proportioning monitoring and adjusting system. BACKGROUND

[0002] Iron smelting material proportioning is a key link in the iron smelting process and directly affects the quality and performance of iron, but at present, the iron smelting material proportioning is mainly calculated by manual calculation, which is prone to proportioning calculation errors, leading to unbalanced material proportioning, and due to the fluctuation of raw material components, the components of raw materials such as iron ore and coke are unstable, leading to inaccurate proportioning, which further affects the quality of iron smelting.

[0003] At the same time, there are problems such as uneven mixing of materials, affecting the smelting effect, improper use of additives, affecting the quality of iron, equipment failure leading to inaccurate proportioning or production interruption, etc., thereby affecting the quality and efficiency of iron smelting. SUMMARY

[0004] The purpose of the application is to solve the above technical problems, and the application provides an AI-based iron smelting material proportioning monitoring and adjusting system, which aims to improve the proportioning accuracy of iron smelting materials and ensure the quality and efficiency of iron smelting.

[0005] In some embodiments of the application, an AI agent and multiple iron smelting periods are established based on a central control unit, the AI agent generates an optimal material proportioning strategy according to the iron smelting material parameters in a single iron smelting period, and sets corresponding feeding parameters according to the material proportioning strategy, avoids uneven mixing of materials, and at the same time, through monitoring the real-time feeding amount, timely warning and correction of feeding deviation are carried out, to avoid the influence of material proportioning on the quality of iron smelting.

[0006] In some embodiments of the application, multiple disturbance indexes are established according to the equipment parameters of the direct current electric arc furnace, the material proportioning and feeding parameters are adjusted in real time by monitoring each disturbance index, to avoid the influence of equipment operation fluctuation on the quality of iron smelting, at the same time, the potential operation risk of the equipment is timely warned to ensure the stability of the iron smelting process, avoid problems such as inaccurate proportioning or production interruption caused by equipment failure, and improve the efficiency of iron smelting.

[0007] In some embodiments of the application, an AI-based iron smelting material proportioning monitoring and adjusting system is provided, which comprises:

[0008] A central control unit is used to set multiple monitoring points.

[0009] A material unit is used to collect real-time furnace charging material parameters.

[0010] The monitoring unit includes multiple monitoring sub-modules, which are set at various monitoring points. The monitoring unit is used to collect the operating parameters of the DC submerged arc furnace.

[0011] The central control unit includes:

[0012] The first processing module is used to build the AI ​​agent;

[0013] The first processing module is also used to establish multiple iron smelting cycles, and the AI ​​agent is used to set the material sub-strategy for each iron smelting cycle.

[0014] In some embodiments of this application, the central control unit further includes:

[0015] The second processing module is used to generate a feeding deviation value based on the real-time furnace material parameters, and to determine whether to generate a correction command based on the feeding deviation value.

[0016] The third processing module is used to acquire the monitoring data packets of the monitoring unit, and the AI ​​agent determines whether to generate an adjustment command based on the monitoring data packets.

[0017] In some embodiments of this application, the first processing module is further configured to:

[0018] Establish a series of material characteristic indicators A, A=(a1,a2…a ... i …a n ), where a i Let be the i-th material characteristic index; n is the number of material characteristic indices;

[0019] Quantification strategies for each material characteristic index are generated, and material analysis sub-models are generated based on all quantification strategies.

[0020] Multiple training data packages are established based on historical iron smelting parameters, and a ratio sub-model is generated based on the iterative results of all training data packages.

[0021] An AI agent is established based on the material analysis sub-model and the proportioning sub-model.

[0022] In some embodiments of this application, the first processing module is further configured to:

[0023] Obtain the expected material parameters within the current iron smelting cycle;

[0024] The logistics analysis sub-model generates a material analysis table for the current ironmaking cycle based on expected material parameters.

[0025] The proportioning sub-model generates the material proportioning strategy for the current ironmaking cycle based on the material analysis table;

[0026] Multiple time intervals are set within the current iron smelting cycle;

[0027] establishes a time interval sequence T, T=(t1, t2…t i …t m ), where t i is the i-th time interval; m is the number of time intervals;

[0028] According to the material proportioning strategy, the delivery sub-strategy of each time interval is set.

[0029] The end time node of each time interval is set as the feedback time node.

[0030] In some embodiments of the present application, the second processing module is further configured to:

[0031] Obtain the furnace charging material parameters collected by the material unit at the current feedback time node;

[0032] Set the time interval corresponding to the current feedback time node as the target time interval;

[0033] Obtain the feedback data packet and the delivery sub-strategy of the target time interval;

[0034] Generate the material delivery deviation value f of the current feedback time node according to the delivery sub-strategy and the feedback data packet;

[0035] Pre-set a material deviation value threshold F1;

[0036] If f>F1, generate a correction instruction for the current feedback time node.

[0037] In some embodiments of the present application, generating the material delivery deviation value f of the current feedback time node comprises:

[0038] f=e1*Q1*[ η i *(p i -p' i ) 2 ]+e2*Q2*U;

[0039] Wherein e1 is a pre-set first weight coefficient; e2 is a pre-set second weight coefficient; Q1 is a pre-set first fixed coefficient; Q2 is a pre-set second fixed coefficient; is the number of material categories in the target time interval; η i is the i-th material category in the target time interval; p i is the actual delivery amount of the i-th material category in the target time interval; p' i is the expected delivery amount of the i-th material category in the target time interval; and U is a historical deviation value.

[0040] In some embodiments of the present application, the third processing module is further configured to:

[0041] a plurality of disturbance indexes are set according to historical operation parameters of the direct current smelting furnace;

[0042] a disturbance index sequence B, B = (b1, b2…b i …b r ), is established, where bi is the i th disturbance index; r is the number of disturbance indexes; i

[0043] a first operation value of each disturbance index is set;

[0044] a disturbance sub-model is established according to all the disturbance indexes;

[0045] the disturbance sub-model generates a disturbance evaluation value of each preset feedback time node in a current smelting period;

[0046] whether to generate an adjustment instruction is determined according to the disturbance evaluation value.

[0047] In some embodiments of the present application, when determining whether to generate an adjustment instruction according to the disturbance evaluation value, the following steps are included:

[0048] a monitoring data packet of the monitoring unit at a current feedback time node is obtained;

[0049] a delivery sub-strategy corresponding to a time interval of the current feedback time node is obtained, and a second operation value of each disturbance index at the current feedback time node is generated;

[0050] a disturbance evaluation value g of the current feedback time node is generated according to the monitoring data packet;

[0051] g = e3*Q3 β i *(j i -j' 1i ) 2 + e4*Q4 β i *(j i -j' 2i ) 2 ;

[0052] where e3 is a preset third weight coefficient; e4 is a preset fourth weight coefficient; Q3 is a preset third fixed coefficient; Q4 is a preset fourth fixed coefficient; r is the number of disturbance indexes; βi is an influence factor of the i th disturbance evaluation index; j i is a reference value of the i th disturbance evaluation index generated based on the monitoring data packet of the current feedback time node; j' 1i is the first operation value of the i th disturbance index; j' 2i is the second operation value of the i th disturbance index at the current feedback time node;

[0053] ​A preset disturbance evaluation value threshold G1 is set.

[0054] If g>G1, the current feedback time node generates an adjustment instruction.

[0055] In some embodiments of the present application, the central control unit further comprises:

[0056] A pre-warning module is configured to obtain the charging deviation value and the disturbance evaluation value of all feedback time nodes in the current smelting period;

[0057] A risk evaluation value c of the current smelting period is generated.

[0058] c= (e5*f i +e6*g i );

[0059] Wherein, e5 is a preset fifth weight coefficient; e6 is a preset sixth weight coefficient; m is the number of feedback time nodes in the current smelting period; f i is the charging deviation value of the i-th feedback time node in the current smelting period; g i is the disturbance evaluation value of the i-th feedback time node in the current smelting period.

[0060] A preset risk evaluation value threshold C1 is set.

[0061] If c>C1, the pre-warning module generates a pre-warning instruction.

[0062] In some embodiments of the present application, the central control unit further comprises:

[0063] A fourth processing module is configured to obtain all smelting data in the current smelting period.

[0064] A reinforcement data packet is generated according to all smelting data.

[0065] The fourth processing module is further configured to generate an update instruction of the AI agent according to the reinforcement data packet.

[0066] Compared with the prior art, the smelting material proportioning monitoring and adjusting system based on AI has the following beneficial effects:

[0067] Based on the central control unit, the AI agent and multiple smelting periods are established. The AI agent generates an optimal material proportioning strategy according to the smelting material parameters in a single smelting period, and sets corresponding charging parameters according to the material proportioning strategy, thereby avoiding uneven mixing of materials. At the same time, by monitoring the real-time charging amount, the charging deviation is timely warned and corrected, thereby avoiding the influence of material proportioning on smelting quality.

[0068] According to the equipment parameters of the direct current smelting furnace, a plurality of disturbance indexes are established, by monitoring each disturbance index in real time, timely adjusting the material ratio and feeding parameters, avoiding the influence of equipment operation fluctuation on the smelting quality, at the same time, timely warning the potential operation risk of the equipment, ensuring the stability of the smelting process, avoiding the problems of inaccurate ratio or production interruption caused by equipment failure, and improving the smelting efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 is a structure schematic diagram of an AI-based smelting material ratio monitoring and adjusting system in the preferred embodiment of the present application. DETAILED DESCRIPTION

[0070] The specific embodiments of the present application will be further described in detail below in conjunction with the drawings and examples. The following examples are used to illustrate the present application, but not to limit the scope of the present application.

[0071] In the description of the present application, it should be understood that the terms "center", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0072] The terms "first", "second" are only for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.

[0073] In the description of the present application, it should be noted that, unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection" should be understood in a broad sense, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0074] As shown in Figure 1 , an AI-based smelting material ratio monitoring and adjusting system in the preferred embodiment of the present application comprises:

[0075] A central control unit is used to set a plurality of monitoring points.

[0076] The material unit is used to collect real-time parameters of materials entering the furnace.

[0077] The monitoring unit includes multiple monitoring sub-modules, which are set at various monitoring points. The monitoring unit is used to collect the operating parameters of the DC submerged arc furnace.

[0078] The central control unit includes:

[0079] The first processing module is used to build the AI ​​agent;

[0080] The first processing module is also used to establish multiple iron smelting cycles, and the AI ​​agent is used to set the material sub-strategy for each iron smelting cycle.

[0081] Specifically, the central control unit also includes:

[0082] The second processing module is used to generate a feeding deviation value based on the real-time furnace material parameters, and to determine whether to generate a correction command based on the feeding deviation value.

[0083] The third processing module is used to acquire the monitoring data packets of the monitoring unit. The AI ​​agent determines whether to generate adjustment instructions based on the monitoring data packets.

[0084] Specifically, iron smelting materials include, but are not limited to, iron ore (magnetite Fe3O4, hematite Fe2O3, limonite Fe2O3·nH2O, siderite FeCO3, etc.), alloying elements, fluxes (limestone, dolomite), reducing agents (coke and carbon monoxide), air, etc.

[0085] Specifically, training data packages are generated by collecting historical iron smelting parameters. These packages include the iron smelting quality corresponding to various materials under different proportioning strategies. Based on machine learning, an input-output mapping is established by processing the training data packages. When new raw material parameters are input, the optimal material proportioning strategy can be generated. For example, after inputting parameters such as the type of iron ore, the diameter of the crushed stone, and the impurity content, the optimal proportions of flux, reducing agent, and alloying elements can be output.

[0086] Specifically, the first processing module is also used for:

[0087] Establish a series of material characteristic indicators A, A=(a1,a2…a ... i …a n ), where a i Let be the i-th material characteristic index; n is the number of material characteristic indices;

[0088] Quantification strategies for each material characteristic index are generated, and material analysis sub-models are generated based on all quantification strategies.

[0089] Multiple training data packages are established based on historical iron smelting parameters, and a ratio sub-model is generated based on the iterative results of all training data packages.

[0090] An AI agent is established based on the material analysis sub-model and the proportioning sub-model.

[0091] Specifically, material characteristic indicators include, but are not limited to, multiple parameters such as iron ore type, impurity content, impurity type, fixed carbon content of coke, and CaO content in limestone.

[0092] Specifically, by setting quantitative strategies for various material characteristic indicators, real-time material parameters can be described quickly and accurately, facilitating analysis and processing by AI agents and improving the efficiency of analyzing the proportions of ferrometallurgical materials.

[0093] It is understood that in the above embodiments, an AI agent and multiple iron smelting cycles are established based on the central control unit. The AI ​​agent generates the optimal material ratio strategy based on the iron smelting material parameters in a single iron smelting cycle, thereby improving the accuracy of the iron smelting material ratio.

[0094] In a preferred embodiment of this application, the first processing module is further configured to:

[0095] Obtain the expected material parameters within the current iron smelting cycle;

[0096] The logistics analysis sub-model generates a material analysis table for the current ironmaking cycle based on expected material parameters.

[0097] The proportioning sub-model generates the material proportioning strategy for the current ironmaking cycle based on the material analysis table;

[0098] Multiple time intervals are set within the current iron smelting cycle;

[0099] Establish a time interval sequence T, T=(t1, t2, ..., t3). i …t m ), where t i Let m be the i-th time interval; m is the number of time intervals.

[0100] Set up delivery sub-strategies for each time interval based on the material allocation strategy;

[0101] Set the end time node of each time interval as the feedback time node.

[0102] Specifically, multiple time intervals are established by uniformly dividing a single iron smelting cycle, and the feeding parameters in each time interval are optimized according to the material proportioning strategy set by the AI ​​agent, thereby generating a feeding sub-strategy for each time interval to avoid uneven material mixing.

[0103] Specifically, the second processing module is also used for:

[0104] acquire the parameters of the charging material collected by the material unit at the current feedback time node;

[0105] set the time interval corresponding to the current feedback time node as a target time interval;

[0106] acquire the feedback data packet and the sub-strategy of the target time interval;

[0107] generate a charging deviation value f of the current feedback time node according to the sub-strategy and the feedback data packet;

[0108] preset a material deviation value threshold F1;

[0109] if f>F1, generate a correction instruction at the current feedback time node.

[0110] Specifically, the greater the charging deviation value is, the greater the deviation between the actual amount of the current charging material and the expected amount corresponding to the material ratio strategy is, and the greater the impact on the ironmaking quality is.

[0111] Specifically, the material deviation value threshold F1 can be set according to historical data.

[0112] Specifically, generating the charging deviation value f of the current feedback time node includes:

[0113] f=e1*Q1*[ η i *(p i -p' i )2]+e2*Q2*U;

[0114] wherein e1 is a preset first weight coefficient; e2 is a preset second weight coefficient; Q1 is a preset first fixed coefficient; Q2 is a preset second fixed coefficient; is the number of material categories in the target time interval; η i is an influence factor of the i-th material category in the target time interval; p i is the actual amount of the i-th material category in the target time interval; p' i is the expected amount of the i-th material category in the target time interval; and U is a historical deviation value.

[0115] Specifically, the preset first fixed coefficient and the second fixed coefficient are used to normalize all parameters in the model, so that all parameters in the model are in the same value range.

[0116] Specifically, all material categories in the current ironmaking period are generated by traversing the material ratio strategy, and the actual charging amount of each material list is generated by real-time monitoring of the charging material parameters.

[0117] Specifically, the historical deviation value is generated according to the cumulative value of the charging deviation values of each past feedback time node in the current smelting cycle, and the greater the cumulative value, the greater the corresponding historical deviation value.

[0118] It can be understood that in the above embodiment, the corresponding charging parameters are set according to the material proportioning strategy to avoid uneven mixing of materials, and at the same time, the real-time charging amount is monitored to timely warn and correct the charging deviation, thereby avoiding the influence of material proportioning on smelting quality.

[0119] In the preferred embodiment of the present application, the third processing module is further used for:

[0120] setting a plurality of disturbance indexes according to historical operating parameters of the direct current electric arc furnace;

[0121] establishing a disturbance index sequence B, B=(b1, b2…b i …b r ), wherein b i is the i th disturbance index; r is the number of disturbance indexes;

[0122] setting a first-level operating value of each disturbance index;

[0123] establishing a disturbance sub-model according to all disturbance indexes;

[0124] the disturbance sub-model generates a disturbance evaluation value of each preset feedback time node in the current smelting cycle;

[0125] determining whether to generate an adjustment instruction according to the disturbance evaluation value.

[0126] Specifically, the disturbance indexes include but are not limited to the operating temperature of the boiler, the slag discharge amount, the content of each substance in the slag, and the like.

[0127] Specifically, a plurality of monitoring points are set according to different disturbance indexes, and corresponding data acquisition devices are set according to the types of disturbance indexes required to be collected by each monitoring point.

[0128] Specifically, the monitoring sub-module is preferably various sensors.

[0129] Specifically, the first-level evaluation value of each disturbance index refers to that the current disturbance index is in an optimal state, i.e., the smelting quality is in an optimal state.

[0130] Specifically, the greater the disturbance evaluation value, the greater the deviation between the current smelting state of the boiler and the expected state, and the greater the influence on the smelting quality.

[0131] Specifically, when determining whether to generate an adjustment instruction according to the disturbance evaluation value, it includes:

[0132] The monitoring data packet of the current feedback time node is acquired;

[0133] The sub-dispensing strategy of the current feedback time node corresponding time interval is acquired, and the secondary running value of each perturbation index at the current feedback time node is generated;

[0134] The perturbation evaluation value g of the current feedback time node is generated according to the monitoring data packet;

[0135] g = e3 * Q3 + e4 * Q4 β i *(j i -j' 1i ) 2 + e4 * Q4 β i *(j i -j' 2i ) 2 ;

[0136] Wherein, e3 is a preset third weight coefficient; e4 is a preset fourth weight coefficient; Q3 is a preset third fixed coefficient; Q4 is a preset fourth fixed coefficient; r is the number of perturbation indexes; βi is the influence factor of the i th perturbation evaluation index; j i is the reference value of the i th perturbation evaluation index generated based on the monitoring data packet of the current feedback time node; j' 1i is the primary running value of the i th perturbation index; j' 2i is the secondary running value of the i th perturbation index at the current feedback time node;

[0137] A preset perturbation evaluation value threshold G1 is set;

[0138] If g > G1, the current feedback time node generates an adjustment instruction.

[0139] Specifically, when the perturbation evaluation value exceeds the preset perturbation evaluation value threshold, the real-time deviation of each perturbation index is obtained based on the perturbation instruction according to the perturbation sub-model, and the real-time material ratio and feeding parameters are adjusted to avoid the influence of equipment operation fluctuation on the smelting quality and improve the overall smelting quality. Ensure the safe operation of the equipment.

[0140] Specifically, the preset third fixed coefficient and the fourth fixed coefficient are used to normalize all parameters in the model, so that each parameter in the model is in the same value range.

[0141] In the preferred embodiment of the present application, the central control unit further comprises:

[0142] The early warning module is configured to acquire the feeding deviation value and the perturbation evaluation value of all feedback time nodes in the current smelting period;

[0143] generate a risk evaluation value c of the current smelting period;

[0144] c= (e5*f i +e6*g i );

[0145] wherein e5 is a preset fifth weight coefficient; e6 is a preset sixth weight coefficient; m is a number of feedback time nodes of the current smelting period; f i is a feeding deviation value of the i-th feedback time node in the current smelting period; g i is a disturbance evaluation value of the i-th feedback time node in the current smelting period;

[0146] a preset risk evaluation value threshold C1;

[0147] If c>C1, the warning module generates a warning instruction.

[0148] Specifically, the feeding deviation value and the disturbance evaluation value have the same value range.

[0149] Specifically, the greater the risk evaluation value, the greater the possibility of operation risk of the current direct current electric arc furnace. If the real-time risk evaluation value is greater than the preset risk evaluation value threshold, it indicates that the current direct current electric arc furnace may have an operation failure, and timely maintenance is needed to avoid problems such as inaccurate proportioning or production interruption caused by equipment failure, thereby improving smelting efficiency.

[0150] In the preferred embodiment of the present application, the central control unit further comprises:

[0151] The fourth processing module is configured to acquire all smelting data in the current smelting period;

[0152] generate a reinforcement data packet according to the all smelting data;

[0153] The fourth processing module is further configured to generate an update instruction of the AI agent according to the reinforcement data packet.

[0154] Specifically, all data in a single smelting period are recorded to generate a reinforcement data packet, and the AI agent is iteratively trained using the reinforcement data packet. By periodically iterating the AI agent, the optimization accuracy of material proportioning is enhanced, thereby improving smelting quality and smelting efficiency.

[0155] According to the first concept of the present application, the AI agent and multiple smelting periods are established based on the central control unit. The AI agent generates an optimal material proportioning strategy according to smelting material parameters in a single smelting period, and sets corresponding feeding parameters according to the material proportioning strategy to avoid uneven mixing of materials. At the same time, by monitoring the real-time feeding amount, the feeding deviation is timely warned and corrected to avoid affecting smelting quality due to material proportioning.

[0156] According to the second concept of the present application, a plurality of disturbance indexes are established according to the equipment parameters of the direct current electric arc furnace. By monitoring each disturbance index in real time, the material ratio and the feeding parameters are adjusted in time to avoid the influence of equipment operation fluctuation on the smelting quality. At the same time, the potential operation risk of the equipment is warned in time to ensure the stability of the smelting process, avoid problems such as inaccurate ratio or production interruption caused by equipment failure, and improve the smelting efficiency.

[0157] The above is only the preferred embodiment of the present application. It should be pointed out that for ordinary skilled persons in the technical field, a number of improvements and replacements can be made without departing from the technical principles of the present application, and these improvements and replacements should also be considered as the protection scope of the present application.

Claims

1. An AI-based system for monitoring and adjusting the proportioning of ferrometallurgical materials, characterized in that, include: The central control unit is used to set multiple monitoring points; The material unit is used to collect real-time parameters of materials entering the furnace. The monitoring unit includes multiple monitoring sub-modules, which are set at various monitoring points. The monitoring unit is used to collect the operating parameters of the DC submerged arc furnace. The central control unit includes: The first processing module is used to build the AI ​​agent; The first processing module is also used to establish multiple iron smelting cycles, and the AI ​​agent is used to set the material sub-strategy for each iron smelting cycle. The second processing module is used to generate a feeding deviation value based on the real-time furnace material parameters, and to determine whether to generate a correction command based on the feeding deviation value. The second processing module is also used for: Obtain the furnace feed material parameters collected by the material unit at the current feedback time node; Set the time interval corresponding to the current feedback time point as the target time interval; Obtain feedback data packets and delivery sub-strategies for the target time range; Generate the feeding deviation value f for the current feedback time node based on the feeding sub-strategy and feedback data packet; Preset material deviation threshold F1; If f > F1, a correction instruction is generated at the current feedback time point; Generate the material feeding deviation value f at the current feedback time point, including: f=e1*Q1*[ η i *(p i -p' i ) 2 ]+e2*Q2*U; Where e1 is a preset first weighting coefficient; e2 is a preset second weighting coefficient; Q1 is a preset first fixed coefficient; and Q2 is a preset second fixed coefficient. η represents the quantity of material categories within the target time interval. i p is the influence factor for the i-th material category within the target time interval. i p' represents the actual delivery volume of the i-th material category within the target time interval. i is the expected delivery volume of the i-th material category within the target time interval; U is the historical deviation value.

2. The AI-based ironmaking material proportioning monitoring and adjustment system as described in claim 1, characterized in that, The central control unit also includes: The third processing module is used to acquire the monitoring data packets of the monitoring unit, and the AI ​​agent determines whether to generate an adjustment command based on the monitoring data packets.

3. The AI-based ironmaking material proportioning monitoring and adjustment system as described in claim 2, characterized in that, The first processing module is also used for: Establish a series of material characteristic indicators A, A=(a1,a2…a ... i …a n ), where a i Let be the i-th material characteristic index; n is the number of material characteristic indices; Quantification strategies for each material characteristic index are generated, and material analysis sub-models are generated based on all quantification strategies. Multiple training data packages are established based on historical iron smelting parameters, and a ratio sub-model is generated based on the iterative results of all training data packages. An AI agent is established based on the material analysis sub-model and the proportioning sub-model.

4. The AI-based ironmaking material proportioning monitoring and adjustment system as described in claim 3, characterized in that, The first processing module is also used for: Obtain the expected material parameters within the current iron smelting cycle; The logistics analysis sub-model generates a material analysis table for the current ironmaking cycle based on expected material parameters. The proportioning sub-model generates the material proportioning strategy for the current ironmaking cycle based on the material analysis table; Multiple time intervals are set within the current iron smelting cycle; Establish a time interval sequence T, T=(t1, t2, ..., t3). i …t m ), where t i Let m be the i-th time interval; m is the number of time intervals. Set up delivery sub-strategies for each time interval based on the material allocation strategy; Set the end time node of each time interval as the feedback time node.

5. The AI-based ironmaking material proportioning monitoring and adjustment system as described in claim 4, characterized in that, The third processing module is also used for: Multiple disturbance indicators are set based on the historical operating parameters of the DC submerged arc furnace; Establish a disturbance index sequence B, B=(b1, b2…b i …b r ), where b i Let be the i-th disturbance index; r is the number of disturbance indices; Set the primary operating values ​​for each disturbance index; Establish a perturbation sub-model based on all perturbation indices; The perturbation sub-model generates perturbation evaluation values ​​for each preset feedback time node within the current ironmaking cycle; Whether to generate an adjustment command is determined based on the disturbance evaluation value.

6. The AI-based ironmaking material proportioning monitoring and adjustment system as described in claim 5, characterized in that, When determining whether to generate a control command based on the disturbance evaluation value, the following are included: Obtain the monitoring data packet from the monitoring unit at the current feedback time point; Obtain the delivery sub-strategy for the time interval corresponding to the current feedback time node, and generate the secondary operating values ​​of each disturbance indicator at the current feedback time node; The disturbance evaluation value g for the current feedback time point is generated based on the monitoring data packet; g=e3*Q3* β i *(j i -j' 1i ) 2 ]+e4*Q4* β i *(j i -j' 2i ) 2 ]; Where e3 is the preset third weighting coefficient; e4 is the preset fourth weighting coefficient; Q3 is the preset third fixed coefficient; Q4 is the preset fourth fixed coefficient; r is the number of disturbance indicators; βi is the influence factor of the i-th disturbance evaluation indicator; j i It is the reference value of the i-th disturbance evaluation index generated based on the monitoring data packet at the current feedback time point; j' 1i Let j' be the first-level operating value of the i-th disturbance index; 2i This represents the secondary operating value of the i-th disturbance index at the current feedback time point; Preset disturbance evaluation threshold G1; If g > G1, the adjustment command is generated at the current feedback time node.

7. The AI-based ironmaking material proportioning monitoring and adjustment system as described in claim 6, characterized in that, The central control unit also includes: The early warning module is used to obtain the material feeding deviation value and disturbance evaluation value of all feedback time nodes within the current ironmaking cycle; Generate the risk assessment value c for the current iron smelting cycle; c= (e5*f i +e6*g i ); Where e5 is the preset fifth weighting coefficient; e6 is the preset sixth weighting coefficient; m is the number of feedback time nodes in the current iron smelting cycle; f i This represents the feed deviation value at the i-th feedback time node within the current ironmaking cycle; g i This is the disturbance evaluation value at the i-th feedback time node within the current iron smelting cycle; Preset risk assessment threshold C1; If c > C1, the early warning module generates an early warning instruction.

8. The AI-based ironmaking material proportioning monitoring and adjustment system as described in claim 7, characterized in that, The central control unit also includes: The fourth processing module is used to acquire all iron smelting data within the current iron smelting cycle; Generate an enhanced data package based on all iron smelting data; The fourth processing module is also used to generate update instructions for the AI ​​agent based on the enhanced data package.

Images